1 Senior iOS developer, 1 Middle iOS developer, 1 QA engineer, 1 Project Manager
16 weeks / ~500 man-hours
Waterfall (initial phase), Fixed Price
A US-based digital health startup was developing a custom wearable ECG monitor designed to support both hospital-based and at-home cardiac care. Equipped with Bluetooth-enabled smart sensors, the device captured the electrical signals of a human body and streamed them to a companion iOS application in real time.
The goal? Enable early detection and monitoring of heart conditions through precise data visualization, giving both patients and cardiologists immediate insight into cardiovascular health. The client turned to Expanice to build the mobile layer of this IoT solution—an iOS app capable of rendering high-resolution ECG data live while adhering to the strict data privacy and security principles required for future HIPAA compliance.
Initially, the client came in with a strong technical vision: use an off-the-shelf plotting library to render the ECG signal data in real time. The MVP was scoped as a lightweight visualization app built in Swift, with BLE as the connectivity protocol and Core Plot selected as the graphing engine.
However, early in development, we discovered that the volume and frequency of data produced by the ECG device—300+ data points per second—was far beyond what Core Plot could handle. The visualization lag exceeded acceptable clinical limits, dropping to just 50 points per second.
To solve the challenge, we pivoted to Metal, Apple’s low-level GPU-accelerated graphics API, replacing Core Plot entirely.
Here’s what we did to achieve near-real-time performance:
Even on modern iPhones, rendering 300–500 dots per second with sub-second latency is a serious challenge. Using Metal, we achieved real-time responsiveness by:
This upgrade allowed the app to process and display ECG signals with clinical-grade precision, supporting live diagnostics and session recording.
Like many early-stage startups, the client had a tight budget and fixed timeline. The project kicked off without a discovery phase, which led to underestimating the visualization complexity. Switching from a standard graphing library to Metal mid-project could have caused major scope creep—but our team adapted quickly, staying on budget while meeting all technical benchmarks.
Since then, we’ve made it a best practice to recommend:
While the project was completed as a standalone MVP, it laid the foundation for:
The client successfully proceeded with clinical trials and FDA approval using the application we developed.